Metabotropic glutamate receptor 1 (mGluR1) has been related to processes underlying learning in hippocampal circuits, but demonstrating its involvement in synaptic plasticity when measured directly on the relevant circuit of a learning animal has proved to be technically difficult. We have recorded the functional changes taking place at the hippocampal CA3–CA1 synapse during the acquisition of an associative task in conscious mice carrying a targeted disruption of the mGluR1 gene. Animals were classically conditioned to evoke eyelid responses, using a trace (conditioned stimulus [CS], tone; unconditioned stimulus [US], electric shock) paradigm. Acquisition of this task was impaired in mutant mGluR1+/− mice and abolished in mGluR1−/− mice. A single pulse presented to Schaffer collaterals during the CS–US interval evoked a monosynaptic field excitatory postsynaptic potential at ipsilateral CA1 pyramidal cells, whose slope was linearly related to learning evolution in controls but not in mGluR1 mutants. Long-term potentiation evoked by train stimulation of Schaffer collaterals was also impaired in both mGluR1+/− and mGluR1−/− animals. Administration of the selective mGluR1 antagonist (3aS,6aS)-6a-naphthalen-2-ylmethyl-5-methyliden-hexahydro-cyclopental [c]furan-1-on to wild-type animals mimicked the functional changes associated to mGluR1 insufficiency in mutants. Thus, mGluR1 is required for activity-dependent synaptic plasticity and associative learning in behaving mice.
Metabotropic glutamate receptors (mGluRs) of group I, which comprises mGluR1 and mGluR5, are functionally coupled to the activation of phospholipase C (Pin and Duvoisin 1995; Conn and Pin 1997; Cartmell and Schoepp 2000). They have been linked to different physiological responses to glutamate, including the modulation of pyramidal and nonpyramidal cell excitabilities and synaptic plasticity (Ferraguti et al. 1998, 2004), and are intensely expressed in the hippocampus. In vivo studies have established the involvement of mGluR1 in spatial and associative learning, whereas in vitro studies described its participation in long-term potentiation (LTP; Aiba et al. 1994; Conquet et al. 1994; Mannaioni et al. 2001; Kishimoto et al. 2002; Topolnik et al. 2006). Questions remain, however, regarding how mGluR1 modulates synaptic inputs to CA1 pyramids in alert behaving animals during the actual learning process. Thus, we analyzed here behaving mice with a targeted disruption of the mGluR1 gene to determine their associative learning capabilities in parallel with concomitant changes in the synaptic strength of pyramidal CA3–CA1 synapses.
Hippocampal circuits seem to be involved in different types of associative learning, including classical eyeblink conditioning (Berger et al. 1983; McEchron and Disterhoft 1997; Múnera et al. 2001). Although hippocampal pyramidal cell firing seems to be related to both (delay and trace) classical conditioning paradigms (Múnera et al. 2001), bilateral hippocampal lesions preferentially impair the acquisition of trace (not of delay) conditioning (Thompson 1988; Moyer et al. 1990). It has been shown in humans that trace conditioning requires a declarative (explicit) memory (Eichenbaum 1999) and/or conscious knowledge (Clark and Squire 1998) of established relationships between conditioned stimulus (CS) and unconditioned stimulus (US), a requisite apparently not necessary for the acquisition of delay conditioning. Moreover, mice are capable of acquiring trace conditioning at rates similar to those shown by other mammals (Takatsuki et al. 2003; Domínguez-del-Toro et al. 2004; Gruart et al. 2006).
It has been shown recently in mice and rats that the acquisition of different types of learning is able to modify the synaptic strength of the hippocampal CA3–CA1 synapse (Gruart et al. 2006; Whitlock et al. 2006). In this regard, we were interested in determining the contribution of mGluR1 to both associative learning and the physiological potentiation of CA3–CA1 synapses during the actual acquisition process in conscious mice. For this, we used mGluR1 knockout mice (Conquet et al. 1994) and their littermates as controls. Animals were classically conditioned with a trace paradigm, presenting a tone as a CS and an electrical shock to the supraorbital nerve as an US. Conditioned responses (CRs) were quantified from the electromyographic (EMG) activity of the ipsilateral orbicularis oculi muscle. We also recorded the field excitatory postsynaptic potential (fEPSP) evoked at hippocampal CA1 pyramidal cells by a single pulse presented to the ipsilateral Schaffer collateral–commissural pathway within the CS–US interval. In this way, we were able to follow activity-dependent changes in CA3–CA1 synaptic strength across the successive conditioning sessions (Gruart et al. 2006). Finally, we attempted to determine whether changes in synaptic plasticity during associative learning in mutant mice were accompanied by similar changes in LTP evoked in the CA1 pyramidal area by high-frequency stimulation (HFS) of Schaffer collaterals. The aim was to determine the involvement of mGluR1s in both types of synaptic plasticity in behaving mice. The present results provide convincing evidence of the involvement of hippocampal mGluR1s in both trace eyeblink conditioning and the enhancement in CA3–CA1 synaptic strength concomitant with this type of associative learning. Moreover, LTP was substantially reduced in mGluR1 knockout mice, suggesting that subcellular and molecular processes underlying it share some mechanisms with activity-dependent synaptic plasticity.
Experiments were performed on 106 male adult mGluR1 knockout (−/−), heterozygous (+/−), and control (+/+) mice (Conquet et al. 1994). These animals are proprietary to GlaxoSmithKline (Verona, Italy). To generate the mutant mice, the LacZ gene was inserted into the reading frame of the mGluR1 gene at the level of the second intracellar loop of the seven transmembrane domain. Accordingly, neurons synthesize a fusion protein mGluR1–β-galactosidase that includes the intact N-terminal end of the mGluR1 protein. Mice were maintained in a C57Bl background at the animal facilities of the Servicio de Experimentación Animal, Universidad Miguel Hernández, San Juan de Alicante, Spain. Before surgery, animals were housed in separate cages (n = 10 per cage). Animals were 3–6 months old at the time of surgery, weighing 20–30 g. The mice were kept on a 12-h light/dark cycle with constant ambient temperature (21 ± 1 °C) and humidity (55 ± 9%). Food and water were available ad libitum. Electrophysiological and behavioral studies were carried out in accordance with the guidelines of the European Union (2003/65/CE) and Spanish regulations (BOE 252/34367-91, 2005) for the use of laboratory animals in chronic experiments. Experiments were also approved by the local ethical committees. For classical conditioning, LTP, and pharmacological studies, a minimum of 10 animals were used for each experimental group.
Animals were anesthetized with 1–3% halothane (AstraZeneca, Madrid, Spain) delivered from a calibrated Fluotec 5 (Fluotec-Ohmeda, Tewksbury, MA) vaporizer at a flow rate of 1–4 L/min oxygen. Animals were implanted with bipolar recording electrodes in the left orbicularis oculi muscle and with bipolar stimulating electrodes on the ipsilateral supraorbital nerve (Fig. 2A,B). Electrodes were made of 50 μm, Teflon-coated, annealed stainless steel wire (A-M Systems, Carlsborg, WA). Electrode tips were bared of the isolating cover for ≈0.5 mm and bent as a hook to facilitate a stable insertion in the upper eyelid. Animals were also implanted with bipolar stimulating electrodes in the right (contralateral) Schaffer collateral–commissural pathway of the dorsal hippocampus (2 mm lateral and 1.5 mm posterior to bregma; depth from the brain surface, 1.0–1.5 mm; Paxinos and Franklin 2001) and with a recording electrode in the ipsilateral stratum radiatum underneath the CA1 area (1.2 mm lateral and 2.2 mm posterior to bregma; depth from the brain surface, 1.0–1.5 mm). These electrodes were made of 50 μm, Teflon-coated tungsten wire (Advent Research Materials Ltd, Eynsham, UK). The final position of hippocampal electrodes was determined under recording procedures. A 0.1-mm bare silver wire was affixed to the skull as a ground. The 8 wires were connected to two 4-pin sockets (RS-Amidata, Madrid, Spain). The sockets were fixed to the skull with the help of 2 small screws and dental cement. Further details of implantation procedures have been explained elsewhere (Gruart et al. 2006).
Classical Conditioning Procedures
For classical conditioning recordings, 3 animals at a time were placed in separate small (5 × 5 × 10 cm) plastic chambers located inside a larger Faraday box (30 × 30 × 20 cm). Classical conditioning was achieved using a trace paradigm (Fig. 3A) consisting of a tone (20 ms, 2.4 kHz, 85 dB) presented as a CS. The US consisted of a cathodal, square pulse applied to the supraorbital nerve (500 μs, 3 × threshold) 500 ms after the end of the CS. A total of 2 habituation, 10 conditioning, and 5 extinction sessions were carried out for each animal. A conditioning session consisted of 60 CS–US presentations and lasted ≈30 min. For a proper analysis of the CR, the CS was presented alone in 10% of the cases. CS–US presentations were separated at random by 30 ± 5 s. For habituation and extinction sessions, only the CS was presented, also for 60 times per session, at intervals of 30 ± 5 s.
Electrophysiological Recordings and Stimulation Procedures
Recordings were made using 6 differential amplifiers with a bandwidth of 0.1–10 kHz (P511, Grass-Telefactor, West Warwick, RI). Hippocampal recordings were made with a high-impedance probe (2 × 1012 Ω, 10 pF). As criteria, we considered a “CR” the presence, during the CS–US interval, of EMG activity lasting >10 ms and initiated >50 ms after CS onset. In addition, the integrated EMG activity recorded during the CS–US interval had to be at least 2.5× greater than the averaged activity recorded immediately before CS presentation (Porras-García et al. 2005).
Electrodes were implanted in the CA1 area, using as a guide the field potential depth profile evoked by paired (40 ms of interval) pulses presented to the ipsilateral Schaffer collateral pathway. The recording electrode was fixed at the site where a reliable monosynaptic (≤5 ms) fEPSP was recorded. Synaptic field potentials in the CA1 area were evoked during habituation, conditioning, and extinction sessions by a single 100-μs, square, biphasic (negative–positive) pulse applied to Schaffer collaterals 300 ms after CS presentation. Stimulus intensities ranged from 50 to 400 μA. For each animal, the stimulus intensity was set well below the threshold for evoking a population spike, usually 30–40% of the intensity necessary for evoking a maximum fEPSP response (Gureviciene et al. 2004). An additional criterion for selecting stimulus intensity was that a second stimulus, presented 40 ms after a conditioning pulse, evoked a larger (>20%) synaptic field potential (Bliss and Gardner-Medwin 1973).
For evoking LTP, we used an HFS train consisting of five 200 Hz, 100-ms trains of pulses at a rate of 1/s. This protocol was presented 6 times, at intervals of 1 min. As indicated above for functional synaptic plasticity, pulse intensity (50–400 μA) was set at 30–40% of the amount necessary to evoke a maximum fEPSP response for baseline recordings and after the HFS train. In order to avoid evoking a population spike and/or unwanted electroencephalographic (EEG) seizures, the stimulus intensity during the HFS train was set at the same amount as used for generating the baseline recording.
(3aS,6aS)-6a-naphthalen-2-ylmethyl-5-methyliden-hexahydro-cyclopental [c]furan-1-on (BAY36-7620; Bayer, Wuppertal, Germany) was dissolved in 10% Cremophor (Cremophor EL, Fluka Chemie, Buchs, Switzerland) and 0.9% NaCl. BAY36-7620 was injected at a dose of 20 mg/kg, intraperitoneal (i.p.), 30 min before each conditioning session or before the corresponding LTP study.
At the end of the experiments, mice were deeply reanesthetized (sodium pentobarbital, 50 mg/kg), and perfused transcardially with saline and 4% phosphate-buffered paraformaldehyde (PFA). Selected brain sections (50 μm) including the dorsal hippocampus were obtained in a microtome (Leica, Wetzlar, Germany) and mounted on gelatinized glass slides and Nissl stained with 0.1% toluidine blue, to determine the location of stimulating and recording electrodes.
EMG and hippocampal activity, and 1-V rectangular pulses corresponding to CS and US presentations, were stored digitally on a computer through an analog/digital converter (CED 1401 Plus, Cambridge, UK), at a sampling frequency of 11–22 kHz and an amplitude resolution of 12 bits. Commercial computer programs (Spike 2 and SIGAVG from CED) were modified to represent EMG and fEPSP recordings. Data were analyzed offline for quantification of CRs and fEPSP slope with the help of homemade representation programs (Porras-García et al. 2005; Gruart et al. 2006). Computed data were processed for statistical analysis using the SPSS for Windows package. Unless otherwise indicated, data are represented as the mean ± standard error of the mean. Acquired data were analyzed using a 2-way analysis of variance (ANOVA) test, with group, session, or time as repeated measure. Contrast analysis was added to further study significant differences. Regression analysis was used to study the relationship between the fEPSP slopes and the percentage of CRs.
Animals for immunohistochemistry (n = 10) were deeply anesthetized with ketamine—xylazine, i.p., and perfused transcardially with 4% PFA in 0.12 M phosphate buffer, pH 7.2. Brains were dissected out, postfixed overnight at 4 °C, and stored in PBS. Sections were obtained in a Vibratome (Leica VT1000S) at 100 μm. Sections were blocked in 4% bovine serum albumin, 3% normal horse serum (NHS), 0.2% Triton X-100, and 0.05% azide in phosphate-buffered saline 1× for 2 h. Then, sections were incubated in the primary antibodies diluted in the same blocking solution. We used a polyclonal antibody that recognizes all mGluR1 splice variants (pan-mGluR1 antibody, gift from F. Ciruela, Barcelona, Spain) at a final protein concentration of 2 μg/mL. Other antibodies were a polyclonal antibody to β-galactosidase (Cappel, MP Biomedicals, Irvine, CA; 1:500) and monoclonal antibody Neuronal Nuclei (NeuN; Chemicon, Temecula, CA: 1:1000). Secondary antibodies included several combinations of rabbit anti-mouse IgG AlexaFluor 488 (Molecular Probes, Invitrogen, Carlsbad, CA; 1:500) and goat anti-mouse IgG AlexaFluor 568 (1:500). The secondary antibodies were diluted in blocking solution and applied for 3 h. Sections were then washed in PBS and covered with Citifluor (London, UK). Imaging was performed with a Leica TCS SL confocal microscope.
Expression of mGluR1 in the CA1 Region of the Hippocampus
Both mGluR1 and mGluR5 are intensely expressed in hippocampus CA1 (Ferraguti et al. 1998; Ferraguti and Shigemoto 2006) at postsynaptic and perisynaptic locations (Baude et al. 1993; Lujan et al. 1996). Whereas mGluR5 is abundant in pyramidal cell perikarya and dendrites, expression of the splice variant mGluR1α is mostly confined to interneurons in the alveus and stratum oriens. In order to establish whether an mGluR1 is expressed in CA1, we used a polyclonal antibody that recognizes the N-terminal end of mGluR1, common to all splice variants and conserved in mutant mice (pan-mGluR1 antibody). We detected mGluR1 immunostaining in the stratum pyramidale of CA1 of wild-type, heterozygous, and knockout mice (Fig. 1A,B). The staining was punctate (Fig. 1C,D) and was denser in the superficial tier of the stratum (Fig. 1A–D). Antibodies to mGluR1–β-galactosidase fusion protein in mGluR1+/− or mGluR1−/− mice yielded a similar staining pattern (Fig. 1E–H). CA3 pyramidal cells expressed both pan-mGluR1 and β-galactosidase immunoreactivities (Fig. 1I,J). Our results favor the concept that mGluR1, possibly as mGluR1β (Ferraguti et al. 1998), is expressed not only in CA3 pyramids but also in CA1 pyramids.
EMG and fEPSP Recordings in Behaving Mice
As already shown in a previous study (Gruart et al. 2006), the electrodes implanted in the upper lid did not disturb its normal kinematics and allowed the generation of spontaneous, reflex, and classically conditioned eyelid responses. As a control of the proper location of stimulating and recording electrodes, we checked in all the animals that the electrical stimulation (2 × threshold) of the supraorbital nerve evoked an early EMG activation of the orbicularis oculi muscle at a latency of 4–6 ms, followed by a second EMG activation, with a latency from the stimulus of 15–20 ms (Fig. 2C-2). These successive muscle activations correspond to the R1 and R2 components already described in humans (Kugelberg 1952) and other species of mammals, including mice (Gruart et al. 1995, 2000, 2006; Domínguez-del-Toro et al. 2004). Values collected from mGluR1+/− and mGluR1−/− mice for latency and area of the rectified EMG response for both R1 and R2 components were in the range of those collected from control (mGluR1+/+) animals (P ≥ 0.12, 1-way ANOVA). Thus, the ataxic syndrome characteristic of mGluR1−/− mice (Conquet et al. 1994) did not seem to affect their eyelid motor system (Kishimoto et al. 2002). As a routine check, we confirmed that mGluR1−/− mice were completely unable to perform the Rota-rod test, whereas mGluR1+/− mice performed it as controls did (F4,16 = 1.068; P = 0.317; not illustrated).
The proper location of stimulating and recording electrodes in the dorsal hippocampus is illustrated in Figure 2A. Electrical stimulation of Schaffer collaterals in conscious mice evokes a large negative wave when recorded at the stratum radiatum (i.e., on the apical dendrites of the CA1 pyramidal cells) with a latency of 3.5–4 ms (Figure 2C-1; Gruart et al. 2006). Field EPSPs recorded in the CA1 area occasionally presented a positive shape, suggesting that the recording electrode was located close to the pyramidal cell layer (Schwartzkroin 1986). Field profiles and intensities (in microamperes) for evoking them were similar for the 3 groups of mice.
A spectral analysis of extracellular hippocampal recordings collected from wild-type (+/+), mGluR1+/−, and mGluR1−/− mice indicates no significant differences in their power spectra (see sample records in Fig. 3A; P ≥ 0.16, χ2-distributed test) and the absence of abnormal EEG activities.
Classical Conditioning of Eyelid Responses in Control and mGluR1 Genetically Modified Mice
Thirty-six animals, n = 12 each of wild-type (mGluR1+/+), heterozygous (mGluR1+/−), and knockout (mGluR1−/−) mice, were classically conditioned using a trace (CS, tone; US, shock) paradigm (Fig. 3A). The time interval between the end of the CS and the beginning of the US was 500 ms. The experimental design included the presentation of a single electrical pulse to Schaffer collaterals 300 ms after CS presentation.
In Figure 3A are illustrated single records of the EMG activity of the orbicularis oculi muscle and of the fEPSP evoked in a well-trained animal from each experimental group during the ninth conditioning session. Although all the animals presented normal reflex responses to US presentations, mGluR1−/− mice failed to evoke CRs at the same rate as controls, whereas mGluR1+/− mice presented intermediate values. As quantified in Figure 3D, the learning curve presented by wild-type animals was similar to that in previous descriptions in mice, using similar trace conditioning procedures (Takatsuki et al. 2003; Domínguez-del-Toro et al. 2004; Gruart et al. 2006). Controls presented a mean of 38.5 ± 4% CRs during the first conditioning session and reached asymptotic values from the seventh session onward (>75% of CRs). When compared, the mean percentage of CRs collected from wild-type animals was significantly larger than the corresponding values collected from mGluR1+/− and mGluR1−/− groups across conditioning and extinction sessions. Specifically, differences between controls and mGluR1−/− mice were statistically significant from the third to the 10th conditioning sessions and for all extinction sessions (F32,252 = 12.974; P < 0.001), whereas differences between control and mGluR1+/− groups were significant from the fifth to the 10th conditioning sessions and from the first extinction session (F32,252 = 12.974; P ≤ 0.01). Differences between mGluR1+/− and mGluR1−/− groups were significant from the third to the 10th conditioning sessions and for all extinction sessions (F32,252 = 12.974; P ≤ 0.01). In accordance, wild-type (+/+) mice acquired CRs at a higher rate and to a larger asymptotic value than did mGluR1−/− animals, with the mGluR1+/− group presenting intermediate values.
Evolution of CA3–CA1 Synaptic Field Potentials across Associative Learning
Field EPSPs evoked in wild-type mice by the electrical stimulation of Schaffer collaterals increased progressively in slope (taking the slope of fEPSPs collected during the 2 habituation sessions as 100%) across conditioning sessions to a maximum of ≈125% during the 8th–10th sessions. During extinction, the fEPSP slope decreased to a minimum of 98.5% for the fifth session. In contrast, fEPSP values collected from mGluR1+/− and mGluR1−/− mice reached significantly lower values than controls. In fact, the difference in fEPSP slope between controls and mGluR1−/− mice was statistically significant for all the conditioning and extinction sessions (F32,252 = 4.161; P < 0.001), whereas differences between controls and mGluR1+/− mice were significant from the fifth to the 10th conditioning sessions and from the first to the second extinction sessions (F32,252 = 4.161; P ≤ 0.05). Finally, the differences in fEPSP slope across conditioning between mGluR1+/− and mGluR1−/− groups were significant from the eighth to the 10th conditioning sessions and from the first to the second extinction sessions (F32,252 = 4.161; P ≤ 0.02).
Overall, homozygous (−/−) mGluR1 knockout mice presented an evident impairment for the acquisition of associative learning, together with a noticeable deficit in CA3–CA1 functional synaptic plasticity. In the acquisition of CRs, the mGluR1+/− group presented intermediate values between the wild-type and mGluR1−/− groups, with a significant (P < 0.001) enhancement of CA3–CA1 fEPSPs during the last 2 conditioning sessions (ninth and 10th) as compared with values collected during the 2 habituation sessions.
Relationships between fEPSP Slope and the Percentage of CRs for Controls and mGluR1-Modified Mice
As illustrated in Figure 4, the slope of fEPSPs evoked by Schaffer collaterals stimulation at the CA3–CA1 synapse was linearly related (r ≥ 0.68; P < 0.001) to the percentage of CRs across conditioning (slope = 0.69) and extinction (slope = 0.64) sessions but not during habituation (r = 0.03; P = 0.872). As already suggested (Gruart et al. 2006), the fact that the slopes of the 2 regression lines were similar indicates that the activity-dependent plasticity at the CA3–CA1 synapse could function as a continuum for both acquisition and extinction processes during associative learning. In contrast, no significant linear relationships could be established between fEPSP evolution across conditioning and the increase in the number of CRs per session for both mGluR1+/− (r ≤ 0.12; P ≥ 0.196) and mGluR1−/− (r ≤ 0.25; P ≥ 0.005) mice.
Characteristics of LTP Evoked in Controls and mGluR1-Modified Mice
We tested wild-type and mGluR1+/− and mGluR1−/− mice for enhancement of synaptic transmission using the double-pulse test, at intervals ranging from 10 to 500 ms. It is generally accepted that the facilitation evoked by the presentation of a pair of pulses is a typical presynaptic short-term plastic property of some excitatory synapses of the hippocampus, including the CA3–CA1 synapses, and it has been correlated with neurotransmitter release (Zucker and Regehr 2002). The 3 groups of animals presented a significant (P < 0.01, ANOVA) increase in response to the second pulse at short time intervals (20–40 ms), with no significant differences between groups (P ≥ 0.19, ANOVA; not illustrated). As already reported (Aiba et al. 1994), these results indicate that mGluR1 receptors seem not to be involved in this form of very short-term memory.
To further investigate the deficit in synaptic plasticity observed in mGluR1-modified mice at CA3–CA1 synapses, we decided to carry out a comparative study of the effects of an HFS train presented to Schaffer collaterals in wild-type, mGluR1+/−, and mGluR1−/− mice (n = 10, each; Fig. 5). For baseline records, Schaffer collaterals were stimulated every 5 s for 15 min. The stimulus consisted of a 100 μs, square, biphasic pulse. After HFS, the same single stimulus was presented at the initial rate (12/min) for another 120 min (Fig. 5B). Using this protocol, we found that LTPs evoked were significantly different in controls and mGluR1+/− and mGluR1−/− mice (F134,1206 = 2.107; P < 0.001). Thus, the control group presented an LTP response of >200% of baseline values 15–30 min after HFS (P < 0.01; Fig. 5C). The LTP response of the control group was still significantly larger than baseline values 2 h after the HFS train (P < 0.01; Fig. 5C). In contrast, mGluR1−/− mice showed a limited capability to evoke LTP at the CA3–CA1 synapse. For example, the increase in fEPSP slopes 13–30 min after the HFS train in mGluR1−/− mice was not significantly larger (≈150%; P = 0.12) than for values collected from baseline records (Fig. 5C).
When quantified for comparative purposes 15–30 and 105–120 min after HFS (Fig. 5C), the LTP evoked in the wild-type group was significantly larger than that in mGluR1−/− mice (F2,18 = 10.381; P < 0.01). LTP values collected from the +/+ group were also significantly larger than values collected from the mGluR1+/− group at 15–30 min (210 ± 18% and 150 ± 20%, respectively; P < 0.01) but not at 105–120 min (220 ± 25% and 170 ± 22%, respectively; P = 0.09) after the HFS train. In short, and as already described for activity-dependent synaptic plasticity, the LTP evoked at the CA3–CA1 synapse in controls was significantly larger than the corresponding LTP values collected from mGluR1+/− and mGluR1−/− mice.
Effects of BAY36-7620 on the Acquisition of CRs, Activity-Dependent Synaptic Plasticity, and LTP Evoked at the CA3–CA1 Synapse
The above results prompted us to check whether the pharmacological blockage of mGluR1 receptors in adult wild-type mice would be able to reproduce, at least partially, the deficits in associative learning and in activity-dependent synaptic plasticity observed in genetically modified mice. For this aim, we used BAY36-7620, a potent, selective, and noncompetitive mGluR1 antagonist and inverse agonist (Müller et al. 2000; Carroll et al. 2001). As illustrated in Fig. 6B, animals (n = 10) injected daily (20 mg/kg, i.p., 30 min before each conditioning session) with BAY36-7620 were unable to acquire a classical eyeblink conditioning as controls (n = 10) did (F16,144 = 8.36; P < 0.01). In fact, BAY36-7620–injected animals were unable to evoke more than 40% of CRs, whereas control mice reached some 75% of CRs by the seventh conditioning session. Moreover, BAY36-7620–injected animals presented a small, nonsignificant potentiation of fEPSPs evoked at the CA3–CA1 synapse across conditioning sessions (a maximum of 110% during the eighth session; P = 0.13). The slopes of fEPSPs evoked at the CA3–CA1 synapse in the control group were significantly larger than the corresponding values collected from BAY36-7620–injected animals from the fourth to the 10th conditioning sessions and from the first to the second extinction sessions (F16,144 = 3.153; P < 0.01; Fig. 6A).
In a final group of experiments, we compared LTP evoked in controls (n = 10) and in BAY36-7620–injected animals (n = 10). For this, we used the same protocol described above. As illustrated in Figure 6C,D, both control and BAY36-7620–injected animals presented a significant LTP at the CA3–CA1 synapse following electrical HFS of the ipsilateral Schaffer collaterals (F67,603 = 5.378; P < 0.01; Fig. 6C). However, whereas the control group presented an LTP that reached values >200% of baseline 20 min after HFS, the increase in BAY36-7620–injected animals was only ≈150%. A quantitative analysis of LTP reached in the 2 experimental situations indicated that the LTP evoked in controls was significantly larger than that evoked in BAY36-7620–injected animals (F2,4 = 6.354; P < 0.01; Fig. 6D).
We have shown here in behaving mice that mGluR1 is involved in the acquisition of a trace conditioning paradigm, a conditioning test requiring the participation of hippocampal circuits (Berger et al. 1983; McEchron and Disterhoft 1997; Clark and Squire 1998; Múnera et al. 2001; Gruart et al. 2006). We have also shown that the CA3–CA1 synapse undergoes a slow, progressive increase in fEPSP slopes across conditioning sessions in wild-type animals, but not in mGluR1−/− mice, with mGluR1+/− mice presenting intermediate values. These results indicate that mGluR1 is necessary both for the acquisition of hippocampally dependent trace conditioning and for the proper enhancement in synaptic strength taking place in hippocampal circuits across conditioning sessions (Gruart et al. 2006). The fact that LTP evoked in CA1 pyramidal cells by HFS of Schaffer collaterals was significantly reduced in mGluR1 knockout mice indicates that LTP phenomena share some molecular and cellular mechanisms with the physiological synaptic plasticity evoked during actual associative learning and that mGluR1 is actively involved in both processes.
The mGluR1 Couples Associative Learning and Activity-Dependent Changes at the CA3–CA1 Synapse in Alert Behaving Mice
It is generally assumed that trace conditioning is an associative test involving the participation of hippocampal circuits. In fact, it has been convincingly demonstrated that hippocampal circuits are involved in the acquisition and/or storage of classical eyeblink conditioning (Berger et al. 1983; Moyer et al. 1990). Moreover, unitary recordings of neuronal activity in behaving rabbits and cats have shown that pyramidal cell firing to CS presentations increases across successive conditioning sessions (McEchron and Disterhoft 1977; Múnera et al. 2001; McEchron et al. 2003). This increase in pyramidal cell firing to CS presentations could be directly related to an enhancement of synaptic transmission in the hippocampal circuit, an event demonstrated in vitro at the CA3–CA1 synapses immediately after training (Power et al. 1997), and confirmed recently during in vivo recordings in behaving mice (Gruart et al. 2006).
A previous study (Kishimoto et al. 2002) has shown that both mGluR1−/− mice and mGluR1-rescue mice (which were engineered to reexpress mGluR1 in cerebellar Purkinje cells only) present a severely impaired acquisition of classical conditioning using a trace paradigm, with CS–US intervals (250 or 500 ms) similar to that used here. That study thus established the involvement of mGluR1 in trace conditioning. The present results confirm this finding and extend it by showing that the acquisition of the associative task correlates with a concomitant potentiation of synaptic transmission taking place in the hippocampal circuit. The fact that the increase in CRs across conditioning is linearly related to a similar increase in fEPSP slope at the CA3–CA1 synapse reinforces this contention, mainly because this linear relationship is disrupted in mGluR1 knockout mice. However, other receptors present in the hippocampal circuit could also be involved in the regulation of this activity-dependent synaptic plasticity, as recently demonstrated in behaving mice for TrkB (Gruart et al. 2007) and TrkC (Sahún et al. 2007) receptors, or during spatial learning tasks in mice lacking GluR1 (Reisel et al. 2002).
Pharmacological Blockade of mGluR1 Mimics the Deficits in Learning and Synaptic Plasticity Caused by the Genetic Disruption of mGluR1
We have been able here to replicate the learning and synaptic plasticity deficits observed in mGluR1+/− and mGluR1−/− mice by treating wild-type animals with BAY36-7620, a specific, noncompetitive mGluR1 antagonist with inverse agonist activity (Müller et al. 2000; Carroll et al. 2001). Early in vitro studies rendered contradictory results regarding the effects of group I mGluR antagonists such as (+)-α-methyl-4-carboxyphenylglycine (MCPG); see Bordi et al. (1997). Thus, Bashir et al. (1993) reported that MCPG blocked LTP induction in hippocampal slices, but those results were disputed by others (Manzoni et al. 1994). Nevertheless, more recent studies have convincingly shown that the selective mGluR1 antagonist (S)-(+)-α-amino-4-carboxy-2-methylbenzene-acetic acid (LY 367385) reduced both induction and expression of LTP (Naie and Manahan-Vaughan 2005), as well as reducing kindling-evoked LTP enhancement (Nagaraja et al. 2005)—both in the CA1 region in hippocampal slices—and decreased the spontaneous firing rates of pallidal neurons in awake monkeys (Kaneda et al. 2005). In contrast, the administration of (RS)-3,5-dihydroxyphenylglycine (DHPG), a selective agonist of group I mGluRs, increased pallidal neuron firing, also in awake monkeys (Kaneda et al. 2005). Thus, our finding that BAY36-7620 affects the acquisition of a classical eyeblink conditioning, and impairs the accompanying changes at the CA3–CA1 synapse, strongly supports a definite role of the mGluR1 receptor in activity-dependent synaptic processes underlying associative learning and LTP induction in that hippocampal circuit.
Are LTP and Functional Synaptic Plasticity Related Phenomena?
Present results suggest that both the physiological synaptic potentiation at the hippocampal CA3–CA1 synapse normally accompanying classical eyelid conditioning (Gruart et al. 2006) and LTP evoked experimentally at the same synapse are dependent on mGlu1 receptors. Thus, LTP and activity-dependent synaptic plasticity could be related phenomena. In support of this contention, it has been shown recently that evoking LTP in the hippocampal circuit prevents the acquisition of associative learning (Gruart et al. 2006) and that abolishing LTP after learning is able to erase what was learned (Pastalkova et al. 2006). LTP, however, is a fast increase in synaptic strength produced by HFS of selected pathways, whereas associative learning and the concomitant potentiation in synaptic activities are slow processes requiring a large number of paired CS–US presentations (Woody 1986; Gruart et al. 2007). In any case, LTP shares many functional properties with the physiological increase in synaptic strength accompanying associative learning as input specificity (i.e., the increase in synaptic strength is restricted to the set of synapses in which LTP is evoked) and cooperativity (i.e., a weak input can be potentiated if activated simultaneously with a strong stimulus), but whereas LTP decays across time, activity-dependent synaptic plasticity apparently increases (McNaughton et al. 1978; Levy and Steward 1983; Bliss and Collingridge 1993; Gruart et al. 2006, 2007). The present results indicate that mGluR1 plays a necessary and potentiating role both in the physiological modification of synaptic strength underlying associative learning and in experimentally evoked LTP.
Previous results collected from hippocampal slices of genetically modified mice were contradictory regarding the contribution of mGluR1 to LTP evoked at the CA3–CA1 synapse, reporting either a decrease (Aiba et al. 1994) or no effect (Conquet et al. 1994) in the absence of the receptor. It should be taken into account that in vitro preparations remove many intrinsic circuits involving mGluR1-related neurons (Ferraguti et al. 1998, 2004; Ferraguti and Shigemoto 2006) controlling pyramidal cell excitability (Topolnik et al. 2006). For example, it has been shown in vivo that mGluR1-mutant mice present a reduced LTP in the dentate gyrus as compared with values collected from control wild-type littermates, a fact not detectable during in vitro experiments (Bordi 1996). Taken together, these contradictory results reinforce the need for studying activity-dependent changes in synaptic functional capabilities during the actual learning process in behaving animals.
mGluR1 Contribution to Neural Mechanisms Underlying Synaptic Plasticity and Associative Learning
The location of mGluR1 within the hippocampal circuit is still a matter of debate. The expression of the mGluR1α isoform seems to be restricted to hippocampal interneurons located preferentially (but not exclusively) in the stratum oriens/alveus, whereas the mGluR1β splice variant is expressed in CA3 hippocampal pyramidal neurons and in granule cells of the dentate gyrus (Ferraguti et al. 2004; Ferraguti and Shigemoto 2006). Here, we propose that some mGluR1 receptors may be located postsynaptically on CA1 pyramidal neurons. Indeed, the activation of mGluR1s has noticeable direct excitatory effects on CA1 pyramidal cells, including depolarization and an increase in cell firing (Desai and Conn, 1991; Mannaioni et al. 2001). These effects can be activated by selective agonists (DHPG) of group I mGluRs and blocked by selective antagonists (LY 367385) of mGluR1 (Mannaioni et al. 2001). However, mGluR1 is also involved in the activation of GABAergic hippocampal interneurons, by the modulation of AMPA-receptor–mediated dendritic Ca2+ signals (Topolnik et al. 2005). Finally, it has also been proposed that presynaptic inhibitory effects observed at the CA3–CA1 synapse are mediated by mGluR1s (Mannaioni et al. 2001). In the absence of a more complete information regarding the location and (different) functional roles of mGluR1 within hippocampal circuits, the definite effects of mGluR1 on associative learning, functional synaptic plasticity, and experimentally evoked LTP reported here can be tentatively ascribed to a direct postsynaptic effect of mGluR1 on CA1 pyramidal neurons, to presynaptic effects on CA3 axon terminals (see Cartmell and Schoepp 2000), or to more complex indirect effects involving the participation of hippocampal (mostly inhibitory) interneurons (Bordi et al. 1997). It has also been proposed that group I mGluRs play a homeostatic role in enhancing or preventing the extracellular levels of glutamate (Rodríguez-Moreno et al. 1998; Cartmell and Schoepp, 2000), a role that is evidently distorted in mGluR1 knockout mice.
Spanish Ministry of Education and Science (BFU2004-04660 to A.F., BFU2005-01024 to J.M.D.-G., and BFU2005-02512 to A.G.). Generalitat Valenciana, Spain (CTBPRB/2003/156 to C.G.-S.).
We thank GlaxoSmithKline for breeding pairs of mGluR1 knockout mice; Dr H. Meier of Bayer Health Care AG, Elberfeld, Germany, for kindly providing BAY36-7620 and protocols; Dr F. Ciruela (Barcelona, Spain) for antibodies; Miss María Esteban and Mr Juan Antonio Martínez for technical assistance; and Mr Roger Churchill for his editorial help. Conflict of Interest: None declared.